Anthropic users across online forums are raising the same complaint: Claude suddenly feels… bad. Why it matters: The backlash lands just as Anthropic is testing a more powerful model, Mythos — raising questions about whether cutting-edge AI is becoming less accessible even as it gets more capable. Driving the news: Over the past few weeks, users on X, GitHub and Reddit have been swapping anecdotes, benchmarks and prompts in an effort to pinpoint what changed and why. - "Claude has regressed to the point it cannot be trusted to perform complex engineering," an AMD senior director wrote in a widely shared post on GitHub. - Others have posted side-by-side outputs and benchmarks they say show Claude generating answers that are less…
A manic new phase of the AI boom is sweeping through Silicon Valley, powered by autonomous "agents" capable of liquefying weeks of manual labor into minutes. Why it matters: For now, the frenzy is largely confined to software engineering . But inside that bubble, the shift feels seismic — deepening the gulf between AI builders and bystanders. - "I've followed tech for 25 years and I've never felt a larger gap between the ~1 million people using Codex/Claude and the rest of humanity," tweeted James Wang, director of product marketing at Cerebras. The big picture: Anthropic CEO Dario Amodei recently described the current state of software engineering as the "centaur phase" — a reference to the half-human, half-horse creature of Greek…
Tech workers are increasingly worried that the artificial intelligence they are building will replace them. But some are optimistic that it is just one more tool to work with.
Dan Cox, Axios' chief technology officer, isn't a journalist, but his AI insights are vital to our readers. We want to show, not just tell, the impact AI is having on companies — in this case, Axios. - So we asked Dan to share what it's like to live the AI revolution. What his team is experiencing will soon hit lawyers, marketers, accountants, consultants ... and journalists. Here's what Dan told us: One of our best engineers recently completed a project similar to one he delivered a year ago. - Last year, it took three weeks. This past week, he used AI-based "agent teams" and completed the same amount of work in 37 minutes. Why it matters: This is the reality of the AI hype you read and hear about. Yes, coding is the first big place…